<h1 id="bxr-t2i-upscaler-and-color-grade-(v3-refresh)">BXR - T2I Upscaler &amp; Color Grade (V3 Refresh) 👑</h1>Welcome to the official V3 refresh of the BXR - T2I Upscaler &amp; Color Grade workflow! This is the core foundation I have used from day one, now polished and updated with some slight changes to make it even more robust, efficient, and hardware-friendly.This setup is engineered to generate highly detailed, crisp imagery from scratch, utilizing an advanced multi-stage upscaling pipeline and precision color grading to give your final renders a professional, cinematic pop.<h3 id="what's-new-in-v3">🚀 What's New in V3</h3><ul><li>Integrated VRAM Purging: Added <code>Easy Clean VRAM Used</code> nodes right after the VAE decode to clear GPU memory before the heavy upscale, preventing OOM crashes on standard consumer hardware.</li><li>Refined Ultimate SD Upscale Pipeline: Calibrated to a 3.0x scale using <code>4x_NMKD-Siax_200k.pth</code> with Chess tiling (1536x1536) and a 48-pixel mask blur for seamless, artifact-free high-res outputs.</li><li>Hardware-Accelerated Color Grading: Swapped prompt-based contrast tweaking for the <code>FL Image Adjuster</code> node, locking in rich saturation and contrast (~10.18) for cinematic vibrance.</li><li>Optimized KSampler Routing: Split prompting into base generation and refiner conditioning to isolate structural rendering from texture enhancement.</li></ul><h2 id="workflow-breakdown">⚙️ Workflow Breakdown</h2>The core philosophy of this workflow is balancing high-fidelity generation with optimized VRAM management.[Base Generation] ➔ [VAE Decode / Preview] ➔ [VRAM Purge] ➔ [Ultimate SD Upscale] ➔ [Color Grading Layer]<h3 id="1.-resource-loading-and-lora-management">1. Resource Loading &amp; LoRA Management</h3><ul><li>Load Checkpoint: Built to leverage high-tier realism models (specifically optimized for checkpoints like <code>illustriousRealismBy_v10VAE.safetensors</code>).</li><li>Power LoRA Loader (rgthree): Stacks multiple character, utility, and detail-enhancing LoRAs cleanly without cluttering the workspace. Balances structural enhancements (like <code>DetailerILv2</code>) alongside concept-specific weights.</li></ul><h3 id="2.-latent-and-initial-ksampler">2. Latent &amp; Initial KSampler</h3><ul><li>Empty Latent Image: Starts at a high-resolution base aspect ratio of 1200 x 1600 with a batch size of 3.</li><li>KSampler (Low Res Preview): Runs for 25 steps at a CFG of 6.0. It uses the <code>dpmpp_2m</code> sampler paired with the <code>karras</code> scheduler to lock in the composition and core details efficiently before upscaling.</li></ul><h3 id="3.-ultimate-sd-upscale-pipeline">3. Ultimate SD Upscale Pipeline</h3>The heavy lifting is handled by the Ultimate SD Upscale node, scaling the image up by a massive factor of 3.0x.<ul><li>Denoise Control: Kept low at 0.35 to ensure the upscale strictly enhances details, skin texture, and sharpness without shifting or morphing the original composition.</li></ul><h3 id="4.-post-processing-and-color-grading">4. Post-Processing &amp; Color Grading</h3><ul><li>FL Image Adjuster: Mathematically boosts Saturation and Contrast to eliminate the typical washed-out neural output look, delivering deep, vibrant tones right out of the generator.</li></ul><h2 id="recommended-settings-matrix">📊 Recommended Settings Matrix</h2>Base Generation Stage<ul><li>Steps: 25</li><li>CFG: 6.0</li><li>Sampler: <code>dpmpp_2m</code></li><li>Scheduler: <code>karras</code></li><li>Denoise: 1.00</li><li>Target Dimensions: 1200 x 1600</li></ul>Ultimate SD Upscale Stage<ul><li>Steps: 26</li><li>CFG: 7.0</li><li>Sampler: <code>dpmpp_2m</code></li><li>Scheduler: <code>karras</code></li><li>Denoise: 0.35</li><li>Target Dimensions: 3.0x Scale (~3600 x 4800)</li></ul><h2 id="prompting-strategy-and-guidance">📝 Prompting Strategy &amp; Guidance</h2>This workflow splits prompting into primary generation blocks and specific refiner blocks to maximize control:<ul><li>Primary Prompt: Focus heavily on your core subject matter, clothing, environmental background, specific physical attributes (e.g., maintaining a "Mild" toned level for a character's physique rather than overblown muscle), and lighting moods.</li><li>Refiner Positive Prompt: Do not paste your entire descriptive prompt here! Keep it strictly focused on quality-boosting modifiers, hyper-detailed skin textures, photographic clarity, and eliminating artifacts (e.g., <code>score_9</code>, <code>score_8_up</code>, <code>masterpiece</code>, <code>photorealistic</code>, <code>high-fidelity details</code>).</li><li>Negative Prompts: Standard embedding blocks filter out anatomical mutations, compression artifacts, and poor framing.</li></ul><h2 id="required-custom-nodes">🛠️ Required Custom Nodes</h2>Before running, ensure you have the following custom node suites installed via your ComfyUI Manager:<ul><li><code>rgthree-comfy</code> (Power LoRA Loader)</li><li><code>UltimateSDUpscale</code></li><li><code>ComfyUI-Easy-Use</code> (VRAM purging utilities)</li><li><code>Fill-Nodes</code> (FL Image Adjuster)</li></ul>Note: Verify your models (<code>illustriousRealismBy_v10VAE.safetensors</code> and <code>4x_NMKD-Siax_200k.pth</code>) are placed in their respective <code>checkpoints</code> and <code>upscale_models</code> directories before queuing your first prompt!

BXR - T2I Ultra - Upscale & Color Grade

🚀 Version Update: BXR - T2I Upscaler &amp; Color Grade (v3.0) (BXR Lab Is Back)What's New in This Version:* Integrated VRAM Purging: Added <code>Easy Clean VRAM Used</code> nodes right after the VAE decode to clear GPU memory before the heavy upscale, preventing OOM crashes on consumer hardware.* Refined Ultimate SD Upscale Pipeline: Calibrated to a 3.0x scale using <code>4x_NMKD-Siax_200k.pth</code> with Chess tiling (1536x1536) and a 48-pixel mask blur for seamless, artifact-free high-res outputs.* Hardware-Accelerated Color Grading: Swapped prompt-based contrast tweaking for the <code>FL Image Adjuster</code> node, locking in rich saturation and contrast (~10.18) for cinematic vibrance.* Optimized KSampler Routing: Split prompting into base generation and refiner conditioning to isolate structural rendering from texture enhancement.Required Custom Nodes (Lightweight):* <code>rgthree-comfy</code> (Power LoRA Loader)* <code>UltimateSDUpscale</code>* <code>ComfyUI-Easy-Use</code>* <code>Fill-Nodes</code> (FL Image Adjuster)

Files with SHA256: cd1fb2868873568efb3e7e09ac3116d12f53f637a5eb243e3e22824e225db419